Author

Zilin Xiao

Rice University

h-index 4 View on Semantic Scholar

We have 1 of 8 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Aug 2026

Perception Before Reasoning: Dynamic Latent Reasoning for Video Understanding and Question Answering

DyLaR is proposed, which first grounds a question in a short block of perception latents, and then adaptively decides whether to append reasoning latents before answering, and which improves average accuracy over same-backbone baselines while generating fewer than 20 tokens per query.

Hao-Tian Xia, Zilin Xiao, Junbo Zou et al. · 0 citations